Projeto ATD

2ºsemestre 2020/2021

FCTUC - DEI

Membros:

• Joel Oliveira
• Tomás Mendes

import numpy as np
from sympy import *
import matplotlib.pyplot as plt
import pandas as pd
#import seaborn as sns
#sns.set_style("whitegrid")
import math
from scipy import signal as sig

Ler dados do acelerómetro para variável

(1 variável / experiência )

userA_expB (Experiência B do Sujeito A)

user1_1 = pd.DataFrame(data=np.loadtxt("Datasets/acc_exp01_user01.txt"), columns=['X','Y','Z'])
user1_2 = pd.DataFrame(data=np.loadtxt("Datasets/acc_exp02_user01.txt"), columns=['X','Y','Z'])
user2_1 = pd.DataFrame(data=np.loadtxt("Datasets/acc_exp03_user02.txt"), columns=['X','Y','Z'])
user2_2 = pd.DataFrame(data=np.loadtxt("Datasets/acc_exp04_user02.txt"), columns=['X','Y','Z'])
user3_1 = pd.DataFrame(data=np.loadtxt("Datasets/acc_exp05_user03.txt"), columns=['X','Y','Z'])
user3_2 = pd.DataFrame(data=np.loadtxt("Datasets/acc_exp06_user03.txt"), columns=['X','Y','Z'])
user4_1 = pd.DataFrame(data=np.loadtxt("Datasets/acc_exp07_user04.txt"), columns=['X','Y','Z'])
user4_2 = pd.DataFrame(data=np.loadtxt("Datasets/acc_exp08_user04.txt"), columns=['X','Y','Z'])

Juntar à estrutura com os dados a informação do tempo em minutos de cada experiência

freq = 50 # Hz
T = 1/freq # sec

user1_1['Time (min)'] = np.arange(0, len(user1_1['X']) * T, T)/60
user1_2['Time (min)'] = np.arange(0, len(user1_2['X']) * T, T)/60 
user2_1['Time (min)'] = np.arange(0, len(user2_1['X']) * T, T)/60 
user2_2['Time (min)'] = np.arange(0, len(user2_2['X']) * T, T)/60 
user3_1['Time (min)'] = np.arange(0, len(user3_1['X']) * T, T)/60 
user3_2['Time (min)'] = np.arange(0, len(user3_2['X']) * T, T)/60 
user4_1['Time (min)'] = np.arange(0, len(user4_1['X']) * T, T)/60 
user4_2['Time (min)'] = np.arange(0, len(user4_2['X']) * T, T)/60

Guardar os dados das experiências num ficheiro .dat, para ler mais tarde

users_exp = [user1_1, user1_2, user2_1, user2_2, user3_1, user3_2, user4_1, user4_2]
pd.to_pickle(users_exp, "users.dat")

Verificação dos dados presentes no DataSet, através da função 'info()' das DataFrame da livraria Pandas.

Esta função devolve informações sobre a informação presente na DataFrame. É devolvida uma string com a quantidade de valores presente não nula, a quantidade de valores presente com valor nulo e ainda o tipoda informação presente em cada coluna.

Assim, é possível verificar que o DataSet está bastante completo, sem falhas de informação, e o tipo de dados existente nele. No caso dados do tipo float.

for i in range(len(users_exp)):
    print("//" + "_"*9 + f"Sujeito {i//2+1} Experiencia {i%2+1}" + "_"*9 + "\\\\", end = '\n\n')
    users_exp[i].info()
    print("\n")
    

//_________Sujeito 1 Experiencia 1_________\\

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 20598 entries, 0 to 20597
Data columns (total 4 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   X           20598 non-null  float64
 1   Y           20598 non-null  float64
 2   Z           20598 non-null  float64
 3   Time (min)  20598 non-null  float64
dtypes: float64(4)
memory usage: 643.8 KB


//_________Sujeito 1 Experiencia 2_________\\

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 19286 entries, 0 to 19285
Data columns (total 4 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   X           19286 non-null  float64
 1   Y           19286 non-null  float64
 2   Z           19286 non-null  float64
 3   Time (min)  19286 non-null  float64
dtypes: float64(4)
memory usage: 602.8 KB


//_________Sujeito 2 Experiencia 1_________\\

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 18026 entries, 0 to 18025
Data columns (total 4 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   X           18026 non-null  float64
 1   Y           18026 non-null  float64
 2   Z           18026 non-null  float64
 3   Time (min)  18026 non-null  float64
dtypes: float64(4)
memory usage: 563.4 KB


//_________Sujeito 2 Experiencia 2_________\\

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 16565 entries, 0 to 16564
Data columns (total 4 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   X           16565 non-null  float64
 1   Y           16565 non-null  float64
 2   Z           16565 non-null  float64
 3   Time (min)  16565 non-null  float64
dtypes: float64(4)
memory usage: 517.8 KB


//_________Sujeito 3 Experiencia 1_________\\

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 20994 entries, 0 to 20993
Data columns (total 4 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   X           20994 non-null  float64
 1   Y           20994 non-null  float64
 2   Z           20994 non-null  float64
 3   Time (min)  20994 non-null  float64
dtypes: float64(4)
memory usage: 656.2 KB


//_________Sujeito 3 Experiencia 2_________\\

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17493 entries, 0 to 17492
Data columns (total 4 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   X           17493 non-null  float64
 1   Y           17493 non-null  float64
 2   Z           17493 non-null  float64
 3   Time (min)  17493 non-null  float64
dtypes: float64(4)
memory usage: 546.8 KB


//_________Sujeito 4 Experiencia 1_________\\

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17668 entries, 0 to 17667
Data columns (total 4 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   X           17668 non-null  float64
 1   Y           17668 non-null  float64
 2   Z           17668 non-null  float64
 3   Time (min)  17668 non-null  float64
dtypes: float64(4)
memory usage: 552.2 KB


//_________Sujeito 4 Experiencia 2_________\\

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 15888 entries, 0 to 15887
Data columns (total 4 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   X           15888 non-null  float64
 1   Y           15888 non-null  float64
 2   Z           15888 non-null  float64
 3   Time (min)  15888 non-null  float64
dtypes: float64(4)
memory usage: 496.6 KB

for i in range(len(users_exp)):
    print("//" + "_"*15 + f"Sujeito {i//2+1} Experiencia {i%2+1}" + "_"*15 + "\\\\", end='\n\n')
    print(users_exp[i].describe())
    print("\n")

//_______________Sujeito 1 Experiencia 1_______________\\

                  X             Y             Z    Time (min)
count  20598.000000  20598.000000  20598.000000  20598.000000
mean       0.880673     -0.101714      0.097088      3.432833
std        0.346449      0.340091      0.285555      1.982092
min       -0.647222     -1.209722     -0.676389      0.000000
25%        0.805556     -0.285764     -0.081944      1.716417
50%        0.981945     -0.202778     -0.004167      3.432833
75%        1.019445     -0.031944      0.204167      5.149250
max        1.950000      1.036111      1.269444      6.865667


//_______________Sujeito 1 Experiencia 2_______________\\

                  X             Y             Z    Time (min)
count  19286.000000  19286.000000  19286.000000  19286.000000
mean       0.886662     -0.119086     -0.014771      3.214167
std        0.305887      0.410941      0.240264      1.855844
min       -0.870833     -1.170833     -0.929167      0.000000
25%        0.815278     -0.331944     -0.172222      1.607083
50%        0.968056     -0.262500     -0.058333      3.214167
75%        0.998611     -0.058681      0.073611      4.821250
max        1.986111      1.538889      1.348611      6.428333


//_______________Sujeito 2 Experiencia 1_______________\\

                  X             Y             Z    Time (min)
count  18026.000000  18026.000000  18026.000000  18026.000000
mean       0.810453     -0.147388      0.004393      3.004167
std        0.389891      0.387617      0.368416      1.734601
min       -0.563889     -1.227778     -1.493056      0.000000
25%        0.755903     -0.370833     -0.220833      1.502083
50%        0.945833     -0.265278      0.029167      3.004167
75%        1.009722     -0.105556      0.147222      4.506250
max        1.976389      1.476389      1.333333      6.008333


//_______________Sujeito 2 Experiencia 2_______________\\

                  X             Y             Z    Time (min)
count  16565.000000  16565.000000  16565.000000  16565.000000
mean       0.829062     -0.164487      0.221480      2.760667
std        0.363335      0.409688      0.236887      1.594016
min       -0.594444     -1.388889     -0.898611      0.000000
25%        0.738889     -0.384722      0.077778      1.380333
50%        0.952778     -0.288889      0.169444      2.760667
75%        0.993056     -0.105556      0.283333      4.141000
max        1.984722      1.173611      1.304167      5.521333


//_______________Sujeito 3 Experiencia 1_______________\\

                  X             Y             Z    Time (min)
count  20994.000000  20994.000000  20994.000000  20994.000000
mean       0.878041     -0.094788      0.028095      3.498833
std        0.326701      0.396218      0.254908      2.020197
min       -0.136111     -1.243056     -0.647222      0.000000
25%        0.818056     -0.306944     -0.144444      1.749417
50%        0.986111     -0.227778     -0.041667      3.498833
75%        1.013889     -0.019444      0.098264      5.248250
max        1.995833      1.209722      1.200000      6.997667


//_______________Sujeito 3 Experiencia 2_______________\\

                  X             Y             Z    Time (min)
count  17493.000000  17493.000000  17493.000000  17493.000000
mean       0.851377     -0.033049     -0.005243      2.915333
std        0.341561      0.429473      0.277169      1.683313
min       -1.043056     -1.019445     -0.734722      0.000000
25%        0.769444     -0.286111     -0.180556      1.457667
50%        0.959722     -0.191667     -0.076389      2.915333
75%        1.015278      0.040278      0.055556      4.373000
max        1.969445      1.288889      1.176389      5.830667


//_______________Sujeito 4 Experiencia 1_______________\\

                  X             Y             Z    Time (min)
count  17668.000000  17668.000000  17668.000000  17668.000000
mean       0.867816      0.061476      0.181707      2.944500
std        0.369651      0.368682      0.164949      1.700152
min       -0.211111     -1.030556     -0.773611      0.000000
25%        0.827778     -0.140278      0.088889      1.472250
50%        0.995833     -0.051389      0.156944      2.944500
75%        1.025000      0.087500      0.231944      4.416750
max        1.965278      1.140278      1.331944      5.889000


//_______________Sujeito 4 Experiencia 2_______________\\

                  X             Y             Z    Time (min)
count  15888.000000  15888.000000  15888.000000  15888.000000
mean       0.867916      0.114830      0.121352      2.647833
std        0.372438      0.391490      0.148444      1.528872
min       -0.237500     -1.088889     -0.530556      0.000000
25%        0.811111     -0.087500      0.050000      1.323917
50%        0.993056     -0.015278      0.109722      2.647833
75%        1.029167      0.175347      0.168056      3.971750
max        1.969445      1.323611      1.325000      5.295667

Leitura dos dados inseridos no ficheiro 'users.dat', anteriormente.

Este ficheiro contém um array com as informações de cada experiência inseridas numa DataFrame

users_exp = pd.read_pickle("users.dat")
user1_1, user1_2, user2_1, user2_2, user3_1, user3_2, user4_1, user4_2 = users_exp

Atividades Dinâmicas:

1. WALKING

2. WALKING_UPSTAIRS

3. WALKING_DOWNSTAIRS

Atividades Estáticas:

1. SITTING

2. STANDING

3. LAYING

Atividades de Transição:

1. STAND_TO_SIT

2. SIT_TO_STAND

3. SIT_TO_LIE

4. LIE_TO_SIT

5. STAND_TO_LIE

6. LIE_TO_STAND

Função para escrever manualmente as atividades nos plots

def plot_with_labels(upxvalues, downxvalues, up, down, user, index):
    '''
    Plots the X, Y, Z axis from the accelerometer
            Parameters:
                    upxvalues (list): ordenadas para impressão de up (labels a colocar em cima)
                    downxvalues (list): ordenadas para impressão de down (labels a colocar em baixo)
                    up (list): lista com as labels das atividades
                    down (list): list com as labels das atividades
                    user (pandas.dataFrame): dataFrame com os dados do user a imprimir.
                    index (int): numero da experiência, de modo a imprimir o Numero do Sujeito, bem como da sua experiência
    '''
    figure, subplots = plt.subplots(nrows=3, ncols=1, figsize=(16,8))
    figure.suptitle(f"Valores Obtidos Pelo Acelerómetro na Experiência {index%2+1} do Sujeito {index//2+1}", fontsize = 'xx-large')
    
    #X
    subplots[0].plot( user['Time (min)'], user['X'], lw = 0.3)
    subplots[0].set_xlabel("Time (min)")
    subplots[0].set_ylabel("ACC_X")
    subplots[0].set_xlim(0, max(user['Time (min)']))
    
    #Y
    subplots[1].plot( user['Time (min)'], user['Y'], lw = 0.3)
    subplots[1].set_xlabel("Time (min)")
    subplots[1].set_ylabel("ACC_Y")
    subplots[1].set_xlim(0, max(user['Time (min)']))
    
    #Z
    subplots[2].plot( user['Time (min)'], user['Z'], lw = 0.3)
    subplots[2].set_xlabel("Time (min)")
    subplots[2].set_ylabel("ACC_Z")
    subplots[2].set_xlim(0, max(user['Time (min)']))
    
    xmin,xmax = subplots[0].get_xlim()
    
    for i in range(len(upxvalues)):
        yv = 0.95 if i%2==0 else 0.90
        frase = up[i]
        xv = upxvalues[i]/xmax
        
        subplots[0].annotate(xycoords = 'axes fraction', text = frase, xy = (xv,yv), textcoords = subplots[0].transAxes, xytext = (xv,yv), fontweight='bold')
        subplots[1].annotate(xycoords = 'axes fraction', text = frase, xy = (xv,yv), textcoords = subplots[1].transAxes, xytext = (xv,yv), fontweight='bold')
        subplots[2].annotate(xycoords = 'axes fraction', text = frase, xy = (xv,yv), textcoords = subplots[2].transAxes, xytext = (xv,yv), fontweight='bold')
        
    yv = 0.01 
    for i in range(len(downxvalues)):
        frase = down[i]
        xv = downxvalues[i]/xmax
        
        subplots[0].annotate(xycoords = 'axes fraction', text = frase, xy = (xv,yv), textcoords = subplots[0].transAxes, xytext = (xv,yv), fontweight='bold')
        subplots[1].annotate(xycoords = 'axes fraction', text = frase, xy = (xv,yv), textcoords = subplots[1].transAxes, xytext = (xv,yv), fontweight='bold')
        subplots[2].annotate(xycoords = 'axes fraction', text = frase, xy = (xv,yv), textcoords = subplots[2].transAxes, xytext =(xv,yv), fontweight='bold')
    figure.tight_layout()

Visualização das atividades para todas as experiências do nosso dataset

downintervals = [0.08, 0.44, 0.78,  1.2, 1.6, 1.9, 2.5,  3.2, 4.4, 5, 5.6]
upintervals = [0.39, 0.7, 1.1, 1.5, 1.9, 2.1, 2.8, 3.55, 4.7, 5.25, 5.8  ]

downlabels = ['STAND', 'SIT', 'STAND','LAY', 'SIT', 'LAY', 'W', 'W','W_D', 'W_D', 'W_D']
uplabels = ['STAND_SIT', 'SIT_STAND', 'STAND_LIE', 'LIE_SIT', 'SIT_LIE', 'LIE_STAND', 'W', 'W', 'W_U', 'W_U', 'W_U' ]

plot_with_labels(upintervals, downintervals, uplabels, downlabels, user1_1, 0)

downintervals = [ 0.08, 0.45, 0.75, 1.3, 1.55, 1.95, 2.5, 3.4, 4.4, 5, 5.6]
upintervals = [ 0.40, 0.7, 1.15, 1.5, 1.8, 2.1, 2.9, 3.8, 4.75, 5.3, 5.9 ]

downlabels = [ 'STAND', "SIT", "STAND", "LAY", "SIT", "LAY", "W", "W", "W_D", "W_D", "W_D" ]
uplabels = [ "STAND_SIT", "SIT_STAND", "STAND_LIE", "LIE_SIT", "SIT_LIE", "LIE_STAND", "W", "W", "W_U", "W_U", "W_U" ]

plot_with_labels(upintervals, downintervals, uplabels, downlabels,user1_2, 1)

downintervals = [ 0.12, 0.6, 1, 1.5, 1.85, 2.2, 2.5, 3.4, 4.25, 4.9, 5.4 ]
upintervals = [ 0.45, 0.87, 1.3, 1.6, 2.05, 2.5, 2.9, 3.95, 4.6, 5.15 ]

downlabels = [ "STAND", "SIT", "STAND", "LAY", "SIT", "LIE", "W", "W", "W_U", "W_U", "W_U"]
uplabels = [ "STAND_SIT", "SIT_STAND", "STAND_LIE", "LIE_SIT", "SIT_LIE", "LIE_STAND", "W", "W_D", "W_D", "W_D"]

plot_with_labels(upintervals, downintervals, uplabels, downlabels, user2_1, 2)

downintervals = [ 0.2, 0.6, 0.8, 1.2, 1.5, 1.9, 2.3, 2.95, 3.8, 4.4, 4.9 ]
upintervals = [ 0.4, 0.7, 1.1, 1.4, 1.7, 2.05, 2.6, 3.5, 4.1, 4.65 ]

downlabels = [ "STAND", "SIT", "STAND", "LAY", "SIT", "LIE", "W", "W", "W_U", "W_U", "W_U" ]
uplabels = [ "STAND_SIT",  "SIT_STAND", "STAND_LIE", "LIE_SIT", "SIT_LIE", "LIE_STAND", "W", "W_D", "W_D", "W_D" ]

plot_with_labels(upintervals, downintervals, uplabels, downlabels, user2_2, 3)

downintervals = [ 0.1, 0.55, 0.9, 1.3, 1.7, 2.2, 2.6, 3.6, 4.7, 5.3, 6, 6.6]
upintervals = [ 0.45, 0.7, 1.1, 1.6, 1.95, 2.3, 3, 4.3, 5.05, 5.7, 6.3 ]

downlabels = [ "STAND", "SIT", "STAND", "LAY", "SIT", "LAY", "W", "W", "W_U", "W_U", "W_U", "W_U"]
uplabels = [ "STAND_SIT", "SIT_STAND", "STAND_LIE", "LIE_SIT", "SIT_LIE", "LIE_STAND", "W", "W_D", "W_D", "W_D", "W_D" ]

plot_with_labels(upintervals, downintervals, uplabels, downlabels, user3_1, 4)

downintervals = [ 0.2, 0.7, 1, 1.5, 1.9, 2.3, 2.6, 3.5, 4.3, 4.9, 5.45 ]
upintervals = [ 0.6, 0.8, 1.3, 1.7, 2.1, 2.5, 3, 4, 4.6, 5.2 ]

downlabels = [ "STAND", "SIT", "STAND", "LAY", "SIT", "LAY", "W", "W", "W_U", "W_U", "W_U", "W_U"]
uplabels = [ "STAND_SIT", "SIT_STAND", "STAND_LIE", "LIE_SIT", "SIT_LIE", "LIE_STAND", "W", "W_D", "W_D", "W_D", "W_D" ]

plot_with_labels(upintervals, downintervals, uplabels, downlabels, user3_2, 5)

downintervals = [ 0.15, 0.6, 0.9, 1.3, 1.7, 2.1, 2.45, 3.3, 4.3, 4.9, 5.45 ]
upintervals = [ 0.4, 0.7, 1.1, 1.6, 1.9, 2.3, 2.9, 3.8, 4.6, 5.2 ]

downlabels = [ "STAND", "SIT", "STAND", "LAY", "STAND", "LAY", "W", "W", "W_U", "W_U", "W_U", "W_U"]
uplabels = [ "STAND_SIT", "SIT_STAND", "STAND_LIE", "LIE_STAND", "STAND_LIE", "LIE_STAND", "W", "W_D", "W_D", "W_D", "W_D" ]

plot_with_labels(upintervals, downintervals, uplabels, downlabels, user4_1, 6)

downintervals = [ 0.15, 0.6, 0.9, 1.3, 1.6, 2, 2.4, 3.1, 3.9, 4.4, 4.9 ]
upintervals = [ 0.4, 0.7, 1.1, 1.6, 1.9, 2.2, 2.7, 3.6, 4.2, 4.6 ]

downlabels = ["STAND", "SIT", "STAND", "LAY", "STAND", "LAY", "W", "W", "W_U", "W_U", "W_U", "W_U"]
uplabels = [  "STAND_SIT", "SIT_STAND", "STAND_LIE", "LIE_STAND", "STAND_LIE", "LIE_STAND", "W", "W_D", "W_D", "W_D", "W_D"  ]

plot_with_labels(upintervals, downintervals, uplabels, downlabels, user4_2, 7)